Quantitative Golgi and electron microscopic data suggest a net age-related increase in dendritic extent in regions of the normally aging human, monkey and rat brain that are also losing neurons, which leads us to hypothesize a plastic, compensatory response of surviving neurons to the age-related death of their neighbors. Additional evidence suggests absence of this net age-related increase in dendritic extent in Alzheimer's disease (AD), which leads us to hypothesize a decreased ability of the AD brain to mount a compensatory response. We propose to test these hypotheses by quantification of the growth-associated protein, GAP-43, its isoforms and its message in normally aging and AD human brain. Preliminary data are presented indicating deficiencies in GAP-43 in the AD brain. A cascade of intercellular signaling is proposed which starts with altered expression of proteins by dying neurons -> signals to glia -> altered expression of neurotrophic/toxic and neurite elongation factors by glia -> response of the surviving neurons. We hypothesize that IL-1beta is one of the molecules in this signal cascade. We also suggest that in AD there is a lesion(s) in this cascade. Preliminary data are presented that support the hypothesized relationship between glia and dendritic extent, and a failure of this relationship in early onset AD. Studies are supposed to test selected aspects of this model, as well as where the signaling may fail in AD. As part of these studies we propose to quantify levels of at least one known protein, IL-1beta, and its message in the aging and AD brain, as well as in a microexplant in vitro model system as a source of these molecules. Since no single method of isolation is suitable for all molecules we propose to emphasize peptides since signal molecules in the brain are often peptides (e.g. many peptide transmitter candidates, hormones and trophic molecules). We propose to use reverse phase HPLC, and recent modifications, to isolate these peptides. This is the method of choice for peptide isolation since peptides are usually cleaved from larger precursor molecules by posttranslational processing. Peptides isolated will be screened for activities proposed by the models, and interesting peptides will be sequenced. Data bases will be explored for any match with the sequences developed. Antibodies and cDNA probes to selected peptides will be developed and used to test for the presence of the these peptides and their messages in rodent brain in relation to neuron death and in normally aging and AD brain. Regardless of whether the studies proposed support the hypothesis of growth of neuronal processes in the normally aging brain and relative failure of this growth in the AD brain, they should provide solid data toward evaluating this hypothesis. Study of the proposed model of intercellular signaling will be useful regardless of whether the hypotheses of growth of neuronal processes remain tenable, for it will provide information concerning the mechanisms by which neuron death is associated with gliosis, as well as the production of neurotrophic/toxic and neurite elongation factors by glia, and the effects of AD on selected aspects of these processes. As such, the studies proposed will provide another entry point into study of the pathological cascade that results in Alzheimer's disease.